PEDIATRICS Vol. 108 No. 5 November 2001, pp. 1211-1214

EXPERIENCE AND REASON:
Time Course of Changes in Diffusion-Weighted Magnetic Resonance Imaging in a Case of Neonatal Encephalopathy With Defined Onset and Duration of Hypoxic-Ischemic Insult

	ABSTRACT

Top Abstract Introduction Discussion Conclusion References

The onset and duration of hypoxic-ischemic (HI) insults rarely can be determined precisely in perinatal asphyxia. The need to establish the timing of HI insults will be critical for the successful application of evolving neuroprotective therapies that may be administered to the asphyxiated newborn. Diffusion-weighted magnetic resonance imaging has emerged as an imaging technique that can be used to identify HI brain injury before the detection of abnormalities by conventional magnetic resonance imaging. This case illustrates the early changes in diffusion-weighted and conventional magnetic resonance imaging studies and in quantitative values of the apparent diffusion coefficient in a unique case of neonatal asphyxia in which the onset and duration of the HI insult were known.hypoxia-ischemia, newborn brain, perinatal asphyxia, diffusion-weighted imaging, proton magnetic resonance spectroscopy.

One of the most difficult problems in the evaluation of the newborn with perinatal asphyxia is the determination of the timing in the acute period of the hypoxic-ischemic (HI) insult to the brain. This early determination is becoming increasingly important with the advent of hypothermia and other potential neuroprotective therapies for the treatment of HI brain injury in the asphyxiated newborn. The onset and duration of the HI insult in the vast majority of cases of perinatal asphyxia cannot be reliably determined based on the obstetrical data and clinical course. Studies using computed tomography and conventional magnetic resonance imaging (MRI) have defined the spectrum of lesions observed in the first week of life after perinatal asphyxia. Recently, diffusion-weighted imaging (DWI) has made possible the earlier detection of HI brain injury.¹ This technique has been applied to the human newborn with perinatal asphyxia to demonstrate evidence of HI brain injury before any abnormality could be detected by conventional MRI.^2-4 The case reported here is unique in that a cardiorespiratory arrest occurred in a 1-day-old term infant, and thereby allowed the precise determination of the time course of the changes in the apparent diffusion coefficient (ADC) in the brain after an HI insult.

	CASE DESCRIPTION

An infant boy was born at 41 weeks' gestation to a 20-year-old gravida-1, para-1 mother, who received prenatal care starting at 17 weeks' gestation. The mother's medical history was significant for depression, exercise-induced asthma, and environmental allergies. She took Wellbutrin (buproprion hydrochloride) throughout the pregnancy, and Ritalin (methylphenidate hydrochloride), Benadryl (diphenhydramine HCl), and tetracycline only early in the first trimester. The pregnancy was otherwise uncomplicated. The infant was born by spontaneous vaginal delivery, 12 hours after spontaneous rupture of membranes. The Apgar scores were 8 and 9 at 1 and 5 minutes, respectively. The infant was reported to be well during the first day after birth and breastfed normally.

At 20 hours after birth, the infant was found to be apneic, cyanotic, and asystolic in his crib. He had been placed in his crib approximately 15 minutes earlier, at which time he appeared to be well. Resuscitation was begun immediately, including intubation, ventilation, chest compressions and the administration of epinephrine, atropine, normal saline, and sodium bicarbonate. A heart rate was established 18 minutes after the onset of the resuscitation. An arterial blood gas 30 minutes after initiation of resuscitation showed a pH of 6.68, pCO₂ of 29.5, pO₂ of 517, and CO₂ of 4.2, for which he received additional sodium bicarbonate. He was noted to have generalized tremors, for which he received phenobarbital 20 mg/kg for presumed clinical seizure activity.

Initial examination on transfer to this institution revealed a nondysmorphic, intubated infant with normal anthropometric measurements and general physical examination. The infant was deeply comatose with no response to noxious stimuli and no spontaneous respirations. All brainstem and deep tendon reflexes were absent. He was hypotonic and had no spontaneous or elicited movements. His clinical examination remained unchanged throughout his hospital stay. Arterial blood gas, serum electrolyte, and glucose measurements were all normal during his admission to this institution. Lactic (acid) dehydrogenase and aspartate aminotransferase were elevated at 1134 U/L, and 92 U/L, respectively. Oliguria was noted, with a maximum creatinine of 0.9 mg/dL. At 7 hours after the arrest, the lactate in the cerebrospinal fluid was elevated at 7.8 mmol/L (normal pyruvate of 0.16 mmol/L). Electroencephalograms performed at 9 and 28 hours after the cardiorespiratory arrest showed a very suppressed background and intermittent cerebral activity with an amplitude of 5 to 10 µV. No electrographic seizures were observed. Support was withdrawn 40 hours postarrest and he died immediately thereafter. Permission for autopsy was not granted.

Conventional MRI and line scan diffusion-weighted imaging (LSDI) studies were obtained 6 and 32 hours after the infant's cardiorespiratory arrest. Conventional T1- (TE 14 and TR of 300) and T2-weighted fast spin echo (effective TE 126 and TR of 3000) scans were obtained. LSDI was performed at 10 axial locations with an effective TE of 62.5 ms, TR of 1520 ms, field of view (FOV) 20 × 15 cm, effective section thickness 7 mm, gap 0 mm, and a 128 × 128 acquisition matrix. The b values used were 5 s/mm², and a maximum b value of 750 s/mm², applied along the 3 orthogonal directions. ADC maps were generated off-line by extrapolating to a b value of 1000 s/mm², and ADC values were measured for various regions of interest. Multivoxel proton magnetic resonance spectroscopy (MRS) was performed at the time of the 6-hour scan, and single voxel proton MRS was performed at the follow-up scan 32 hours after the arrest. The multivoxel MRS included mainly basal ganglia, thalamus, and frontal white matter bilaterally and was performed using 2-dimensional chemical-shift imaging with the PRESS technique (TE of 65 ms, TR of 999 ms, FOV 16 × 16 cm, and acquisition matrix of 256 × 192). The single voxel MRS the next day included the left basal ganglia and thalamus. This study was performed using a long-echo PRESS technique with a TE of 144 ms (TR 1500 ms, 256 × 128 matrix, FOV 24 × 24).

At 6 hours after the HI insult, conventional T1- and T2- MRI showed no abnormalities. LSDI showed restricted diffusion (hyperintensity) in the basal ganglia, thalami and dorsal brainstem (Fig 1). ADC values calculated off-line were normal in the cerebral gray and white matter and in the cerebellum, but were decreased in the basal ganglia and thalamus bilaterally, relative to published normative data⁵ (see Table 1). Moreover, multivoxel MRS showed peaks consistent with lactate in the basal ganglia, thalami, and frontal white matter bilaterally (data not shown). At 32 hours after the HI insult, there was hyperintensity in the basal ganglia, thalami, hippocampi, pons, and cerebellum on T2 and proton density MRI. There was T1 hypointensity in the cerebral white matter and some hyperintensity in cerebral gray matter. LSDI showed restricted diffusion with a significant reduction in the ADC values for all regions of interest measured, except for the cerebral white matter (see Fig 1 and Table 1). Single voxel MRS in the left basal ganglia and thalamus showed the presence of a lactate peak with a Lac/Cho (lactate/choline) ratio of 0.38. 

View larger version (112K): [in this window] [in a new window]   Fig. 1.   Images A and B show results of DWI at the level of the basal ganglia and centrum semiovale, respectively, performed 6 hours after cardiorespiratory arrest. Images C and D are at the same levels as in A and B, but were obtained 32 hours after cardiorespiratory arrest. Areas of bright signal indicate areas of restricted diffusion. Note in A the restricted diffusion in basal ganglia (thick arrows) in thalami (curved arrows), and in dorsal brainstem (thin arrows).

	DISCUSSION

Top Abstract Introduction Discussion Conclusion References

This unique case demonstrates the time course of the early changes in the ADC measured by DWI in a case of HI injury in a term newborn in whom the onset and duration of the insult were well-documented. The results of the conventional MRI, DWI and MRS studies illustrate several important points, as follows.

First, despite normal findings on conventional MRI at 6 hours after the insult, the ADC values were clearly decreased in basal ganglia and thalami. After 32 hours, the decreased ADC values were more extensively distributed, and there was further reduction of the ADC in all areas except the frontal white matter. In one study, this finding of a further reduction in ADC and/or enlargement of ADC lesions was observed in several cases of perinatal asphyxia, when a second DWI scan was performed within several days of the first scan.⁴ DWI data in the neonatal rat also have shown a progressive reduction of ADC up to 30 hours after a HI insult, and has been postulated to be attributable to secondary cytotoxic edema, related predominantly to late swelling of glial cells.^6,7

A corollary to these conclusions concerning abnormal ADC values is that the ADC in the cerebral gray and white matter and in the cerebellum was normal 6 hours after a severe HI insult, despite clinical and EEG evidence of diffuse cerebral dysfunction and the subsequent finding of a reduced ADC in the cerebellum and cerebral gray matter. Possible explanations for the lack of ADC abnormality in these areas (in contrast to basal ganglia and thalami) after 6 hours relate to potential differences in the mechanism of cell death, eg, necrosis versus apoptosis, the evolution of delayed cell death, or differences in amounts of extracellular water. Alternatively, the normal ADC values after 6 hours may represent a pseudonormalization of diffusion, as described in rat models of HI insults.^6,8 The study by Rumpel et al showed a biphasic evolution of change in ADC in 7-day-old rats exposed to unilateral HI.⁶ In their study, the ADC decreased markedly as early as 30 minutes after common carotid artery ligation and hypoxia, then normalized by 3 hours, only to decline again beginning about 8 hours after reperfusion was initiated. Moreover, in an earlier study, the same group showed that brief HI insults (15 minutes) may result in a transient reduction in ADC (2 hours post-HI insult) with no evidence of irreversible brain injury by later conventional MRI or histologic examination.⁹ Thus, it is possible that very early after the HI insult DWI scans in the newborn may be misleading, either by the underestimation of severe insults (particularly in the cerebral cortical gray and white matter) or by the detection of transient abnormalities of ADC that may not be indicative of irreversible brain injury.

The MRS studies showed that lactate was detected on both the 6- and 32-hour scans. This finding suggests that brain lactate may be a useful early marker of HI injury. The elevated lactate level in the cerebrospinal fluid (shortly after the 6-hour MRS study) is supportive of this finding. The Lac/Cho ratio that we observed at 32 hours was 0.38, a value high enough to predict a very poor outcome (spastic quadriplegia or death), according to one study of neurologic outcome in neonatal asphyxia by neonatal MRS performed in the first few days of life¹⁰ (normal value of Lac/Cho of 0.02 ± 0.06¹⁰). Thus, the MRS findings were consistent with previously reported data describing the correlation between neonatal MRS and neurologic outcome.

It should be considered whether an elevated lactate is a more reliable indicator of early HI injury than is reduced ADC. As with ADC, there was a biphasic change in lactate after HI insult observed in the newborn piglet.¹¹ However, the lactate did not return to baseline values in the first hours after the HI insult, unlike the change in ADC in rats. Thus, it remains to be determined whether there is a threshold value of Lac/Cho or Lac/Cr measured in the first hours after a HI insult in human newborns that will identify reliably a region at risk for irreversible brain injury.

	CONCLUSION

Top Abstract Introduction Discussion Conclusion References

This case illustrates the time course of early changes in ADC and in metabolites measured by proton MRS after a severe HI insult of known onset and duration in a human newborn. The finding of normal diffusion in the cerebral cortex at 6 hours after the insult suggests that early DWI after HI in human newborns may underestimate the extent and severity of injury, as demonstrated by the follow-up MRI scan at 32 hours after the insult. The lack of restricted diffusion in the cerebral cortex soon after an HI insult may be attributable to pseudonormalization of the ADC as in the animal studies, or it may reflect the slower development of restricted diffusion in cerebral cortex of the human newborn. In this case, MRS appeared to be more valuable than DWI in demonstrating the areas of brain injury in the first hours after the HI insult. Additional studies are needed to elucidate the early evolution of changes in ADC and MRS findings in newborns with asphyxia of variable severity. The determination of the earliest reliable indicators of the extent and severity of HI brain injury will be crucial for the appropriate selection of asphyxiated newborns for neuroprotective therapy.

Janet S. Soul, MDCM, FRCPC^*
Richard L. Robertson, MD
A. Aria Tzika, PhD
Adre J. du Plessis, MBChB, MPH^*
Joseph J. Volpe, MD^*
Departments of ^* Neurology and  Radiology
Children's Hospital, Boston
Boston, MA 02115

	FOOTNOTES

Received for publication Jan 22, 2001; accepted Apr 30, 2001.

Reprint requests to (J.J.V.) Department of Neurology, Fegan 11, Children's Hospital, 300 Longwood Ave, Boston, MA 02115. E-mail: joseph.volpe{at}tch.harvard.edu

	ABBREVIATIONS

HI, hypoxic-ischemic; MRI, magnetic resonance imaging; DWI, diffusion-weighted imaging; ADC, apparent diffusion coefficient; LSDI, line scan diffusion-weighted imaging; FOV, field of view; MRS, magnetic resonance proton spectroscopy; Lac/Cho, lactate/choline.

	REFERENCES

Top Abstract Introduction Discussion Conclusion References

1. Baird AE, Warach S Magnetic resonance imaging of acute stroke [published erratum appears in J Cereb Blood Flow Metab 1998 Oct;18(10):preceding1047]. J Cereb Blood Flow Metab. 1998; 18:583-609 [CrossRef][Medline]
2. Cowan FM, Pennock JM, Hanrahan JD, Manji KP, Edwards AD Early detection of cerebral infarction and hypoxic ischemic encephalopathy in neonates using diffusion-weighted magnetic resonance imaging. Neuropediatrics 1994; 25:172-175 [Medline]
3. Johnson AJ, Lee BC, Lin W Echoplanar diffusion-weighted imaging in neonates and infants with suspected hypoxic-ischemic injury: correlation with patient outcome. AJR Am J Roentgenol 1999; 172:219-226 [Abstract/Free Full Text]
4. Robertson RL, Ben-Sira L, Barnes PD, MR line-scan diffusion-weighted imaging of term neonates with perinatal brain ischemia [see comments]. AJNR Am J Neuroradiol 1999; 20:1658-1670 [Abstract/Free Full Text]
5. Neil JJ, Shiran SI, McKinstry RC, et al. Normal brain in human newborns: apparent diffusion coefficient and diffusion anisotropy measured by using diffusion tensor MR imaging. Radiology. 1998;209;57-66
6. Rumpel H, Nedelcu J, Aguzzi A, Martin E Late glial swelling after acute cerebral hypoxia-ischemia in the neonatal rat: a combined magnetic resonance and histochemical study. Pediatr Res 1997; 42:54-59 [Medline]
7. Nedelcu J, Klein MA, Aguzzi A, Boesiger P, Martin E Biphasic edema after hypoxic-ischemic brain injury in neonatal rats reflects early neuronal and late glial damage. Pediatr Res 1999; 46:297-304 [Medline]
8. Li F, Silva MD, Liu KF, Secondary decline in apparent diffusion coefficient and neurological outcomes after a short period of focal brain ischemia in rats. Ann Neurol 2000; 48:236-244 [CrossRef][Medline]
9. Rumpel H, Buchli R, Gehrmann J, Magnetic resonance imaging of brain edema in the neonatal rat: a comparison of short and long term hypoxia-ischemia. Pediatr Res 1995; 38:113-118 [Medline]
10. Barkovich AJ, Baranski K, Vigneron D, Proton MR spectroscopy for the evaluation of brain injury in asphyxiated, term neonates. AJNR Am J Neuroradiol 1999; 20:1399-1405 [Abstract/Free Full Text]
11. Penrice J, Lorek A, Cady EB, Proton magnetic resonance spectroscopy of the brain during acute hypoxia-ischemia and delayed cerebral energy failure in the newborn piglet. Pediatr Res 1997; 41:795-802 [Medline]